Thymic involution, a co-morbidity factor in amyotrophic lateral sclerosis

Abstract

Amyotrophic lateral sclerosis (ALS) is a devastating disease, characterized by extremely rapid loss of motor neurons. Our studies over the last decade have established CD4(+) T cells as important players in central nervous system maintenance and repair. Those results, together with recent findings that CD4(+) T cells play a protective role in mouse models of ALS, led us to the current hypothesis that in ALS, a rapid T-cell malfunction may develop in parallel to the motor neuron dysfunction. Here, we tested this hypothesis by assessing thymic function, which serves as a measure of peripheral T-cell availability, in an animal model of ALS (mSOD1 [superoxide dismutase] mice; G93A) and in human patients. We found a significant reduction in thymic progenitor-cell content, and abnormal thymic histology in 3-4-month-old mSOD1 mice. In ALS patients, we found a decline in thymic output, manifested in the reduction in blood levels of T-cell receptor rearrangement excision circles, a non-invasive measure of thymic function, and demonstrated a restricted T-cell repertoire. The morbidity of the peripheral immune cells was also manifested in the increase of pro-apoptotic BAX/BCXL2 expression ratio in peripheral blood mononuclear cells (PBMCs) of these patients. In addition, gene expression screening in the same PBMCs, revealed in the ALS patients a reduction in key genes known to be associated with T-cell activity, including: CD80, CD86, IFNG and IL18. In light of the reported beneficial role of T cells in animal models of ALS, the present observation of thymic dysfunction, both in human patients and in an animal model, might be a co-pathological factor in ALS, regardless of the disease aetiology. These findings may lead to the development of novel therapeutic approaches directed at overcoming the thymic defect and T-cell deficiency.

Fig 1

Thymic cell content is reduced in mSOD-1 mice during the course of disease progression. Two-colour flow-cytometry was used to analyse the phenotypes and total cell number in the thymus of mSOD1 mice and their age-matched wild-type controls during the progressive stage of the disease (day 120; A–G) and prior to disease onset (day 60; H). (A–E). Total thymocyte number is significantly reduced at the progressive stage of the disease: (A) Total number of thymocytes (t₁₂=−2.84; *P= 0.01). (B) Total number of DN thymocytes (t₁₂=−2.12; *P= 0.05). (C) Total number of DP thymocytes (t₁₂=−2.97; P=*0.01). (D) Total number of CD4⁺ cells (t₁₂=−2.07; P= 0.06). (E) Total number of CD8⁺ cells (t₁₂=−2.2; P=*0.04). Reduction in thymic cellularity correlated with an increased proportion of DN cells (F, t_6.4= 2.82; *P= 0.02) and decreased proportion of DP cells (G, t_6.6=−2.5; *P= 0.04). (H) Significant elevation in the absolute number of CD8 cells was observed at day 60, prior to disease onset (t_4.23= 2.7; *P= 0.05). Values represent means ± S.E.M. (I) Representative photomicrographs of thymus collected from mSOD-1 mouse at the end-stage of the disease and its age-matched wild-type control. (J) Thymic dysplasia was detected in mSOD1 mice upon disease progression. Sections taken through the thymus were stained by haematoxylin and eosin. (J_i) A representative thymic section derived from a 60-day-old wild-type mouse. Note the distinct separation between the cortex (C) and the medulla (M). (J_ii) Micrograph showing a higher magnification of the indicated area. (J_iii) A representative thymic section derived from 60-day-old mSOD1 mouse, before disease onset. A normal structure is still observed at this time-point. (J_iv) A representative thymic section derived from 120-day-old mSOD1 mouse. No clearly delineated cortex and medulla can be discerned at this stage. Bars represent 100 μm. (K) Vβ T-cell repertoire in the peripheral T-cell pool of mSOD1 and control mice. No major changes in T-cell clonality were observed in mSOD-1 mice compared to age-matched wild-type mice, along disease progression.

Fig 2

Thymic output is severely reduced in ALS patients. Total DNA was purified from PBMCs that were collected from 11 ALS patients (30–62 years old; average age: 49.1 ± 10.3) and 12 age-matched (25–63 years old; average age: 45.1 ± 12.3; P= 0.40) healthy controls. (A) The amount of sjTREC was measured by Q-PCR, and was normalized against RNAseP. A significant reduction was found in the copies of sjTREC in the blood of ALS patients, compared to controls (t_13.4=−3.06; **P= 0.008). The graph shows the interquartile range of the measurements. The dashed line represents the median value; the circle indicates the average value. (B) Scatterplot of sjTREC levels by age.

Fig 3

Normal levels of early T-cell recombination events in ALS patients. Total DNA was purified from PBMCs that were collected from 11 ALS patients (30–62 years old; average age: 49.1 ± 10.3) and 12 age-matched (25–63 years old; average age: 45.1 ± 12.3; P= 0.40) control donors. The early recombination events were quantified using Q-PCR. No differences were found in the two tested events, D-D (A; t₂₁=−0.15; P= 0.88) and D-J (B; t₂₁=−0.35; P= 0.73). The graphs indicate the interquartile range of the measurements. The dashed line represents the median value; the circle indicates the average value.

Fig 4

Restricted TCRγ repertoire in ALS patients. Genescan display of four PCR-generated profiles of TCRγ rearrangements, from five representative patients. Vγ9/2 and Vγfl represent FAM labelled fluorescent variable region primers. Vγ11 and Vγ10/2 are HEX labelled fluorescent variable region primers. Nine additional patients were similarly analysed and demonstrated a similar profile, suggesting restricted polyclonality, though not monoclonality, relative to normal healthy individuals.

Fig 5

Reduced immune potential and increased susceptibility to apoptosis in PBMCs from ALS patients. Expression levels of (A) CD80 (Z= 2.3, *P= 0.02), (B) CD86 (Z= 3.0, **P= 0.002), (C) IL8 (Z=−2.5, *P= 0.01), (D) IFNG (Z= 2.8, **P= 0.006), (E) IL18 (Z= 3.0, **P= 0.002) and (F) BAX to BCL2 ratio (Z=−3.0, **P= 0.002). Expression was assessed by real-time Q-PCR. Values represent expression levels in arbitrary units, normalized against 18S in the same samples (n= 8 and n= 6 for ALS patients and healthy controls, respectively; each was tested in triplicate). Error bars represent S.D.